Good question. In my experience, academics aren't always up to speed on the practicalities of scaling up a lab scale medicine that's been tested on animals and a few humans, to full scale manufacturing and clinical trials and license application. While you can take business risks in doing things in parallel instead of waiting for success in a traditional stage-by-stage process, some things just take time and can't be speeded up. "80% chance of success" does not say ready for all and sundry by the autumn to me.
There is a race to get a vaccine to market as soon as possible. We desperately need one, but we also need it to be safe for as well as for it to work. That is done by clinical trials - and you can't rush these, they take as long as they take. But we need to know if any side-effects are present - if you only test 100 patients, you are not going to pick up side effects that affect 1 in 200 patients. Scale up to 20,000,000 real-life patients with a 1 in 200 side effect and suddenly you have 100,000 people with problems. If the side effect is a bit of a headache for a few hours, then it's not really an issue. But if it's something worse, you could end up with more problems from the vaccine than the disease.
Politicians sticking their oar into this process worries me. Scientists at drug companies and government agencies wrangle over the data for new medicines for months, and the decision to grant a license is always made on the science and on a good balance of benefit to risk. The last thing we need is to rush something out that hasn't been tested enough under pressure from those who don't know any better, only to find that it's doing harm instead.